Merge tag 'amd-drm-next-6.9-2024-03-01' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.9-2024-03-01:

amdgpu:
- GC 11.5.1 updates
- Misc display cleanups
- NBIO 7.9 updates
- Backlight fixes
- DMUB fixes
- MPO fixes
- atomfirmware table updates
- SR-IOV fixes
- VCN 4.x updates
- use RMW accessors for pci config registers
- PSR fixes
- Suspend/resume fixes
- RAS fixes
- ABM fixes
- Misc code cleanups
- SI DPM fix
- Revert freesync video

amdkfd:
- Misc cleanups
- Error handling fixes

radeon:
- use RMW accessors for pci config registers

From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240301204857.13960-1-alexander.deucher@amd.com
Signed-off-by: Dave Airlie <airlied@redhat.com>

11 files changed, 106 insertions, 54 deletions

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index 087d57850304..9e70c41ad98f 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
@@ -2034,6 +2034,31 @@ static int ss_bias_attr_update(struct amdgpu_device *adev, struct amdgpu_device_
 	return 0;
 }
 
+static int pp_od_clk_voltage_attr_update(struct amdgpu_device *adev, struct amdgpu_device_attr *attr,
+					 uint32_t mask, enum amdgpu_device_attr_states *states)
+{
+	uint32_t gc_ver = amdgpu_ip_version(adev, GC_HWIP, 0);
+
+	*states = ATTR_STATE_SUPPORTED;
+
+	if (!amdgpu_dpm_is_overdrive_supported(adev)) {
+		*states = ATTR_STATE_UNSUPPORTED;
+		return 0;
+	}
+
+	/* Enable pp_od_clk_voltage node for gc 9.4.3 SRIOV/BM support */
+	if (gc_ver == IP_VERSION(9, 4, 3)) {
+		if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev))
+			*states = ATTR_STATE_UNSUPPORTED;
+		return 0;
+	}
+
+	if (!(attr->flags & mask))
+		*states = ATTR_STATE_UNSUPPORTED;
+
+	return 0;
+}
+
 /* Following items will be read out to indicate current plpd policy:
  *  - -1: none
  *  - 0: disallow
@@ -2118,7 +2143,8 @@ static struct amdgpu_device_attr amdgpu_device_attrs[] = {
 	AMDGPU_DEVICE_ATTR_RW(pp_sclk_od,				ATTR_FLAG_BASIC),
 	AMDGPU_DEVICE_ATTR_RW(pp_mclk_od,				ATTR_FLAG_BASIC),
 	AMDGPU_DEVICE_ATTR_RW(pp_power_profile_mode,			ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF),
-	AMDGPU_DEVICE_ATTR_RW(pp_od_clk_voltage,			ATTR_FLAG_BASIC),
+	AMDGPU_DEVICE_ATTR_RW(pp_od_clk_voltage,			ATTR_FLAG_BASIC,
+			      .attr_update = pp_od_clk_voltage_attr_update),
 	AMDGPU_DEVICE_ATTR_RO(gpu_busy_percent,				ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF),
 	AMDGPU_DEVICE_ATTR_RO(mem_busy_percent,				ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF),
 	AMDGPU_DEVICE_ATTR_RO(pcie_bw,					ATTR_FLAG_BASIC),
@@ -2163,10 +2189,6 @@ static int default_attr_update(struct amdgpu_device *adev, struct amdgpu_device_
 	} else if (DEVICE_ATTR_IS(pp_dpm_fclk)) {
 		if (mp1_ver < IP_VERSION(10, 0, 0))
 			*states = ATTR_STATE_UNSUPPORTED;
-	} else if (DEVICE_ATTR_IS(pp_od_clk_voltage)) {
-		*states = ATTR_STATE_UNSUPPORTED;
-		if (amdgpu_dpm_is_overdrive_supported(adev))
-			*states = ATTR_STATE_SUPPORTED;
 	} else if (DEVICE_ATTR_IS(mem_busy_percent)) {
 		if ((adev->flags & AMD_IS_APU &&
 		     gc_ver != IP_VERSION(9, 4, 3)) ||
@@ -2174,7 +2196,8 @@ static int default_attr_update(struct amdgpu_device *adev, struct amdgpu_device_
 			*states = ATTR_STATE_UNSUPPORTED;
 	} else if (DEVICE_ATTR_IS(pcie_bw)) {
 		/* PCIe Perf counters won't work on APU nodes */
-		if (adev->flags & AMD_IS_APU)
+		if (adev->flags & AMD_IS_APU ||
+		    !adev->asic_funcs->get_pcie_usage)
 			*states = ATTR_STATE_UNSUPPORTED;
 	} else if (DEVICE_ATTR_IS(unique_id)) {
 		switch (gc_ver) {
diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
index df4f20293c16..eb4da3666e05 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
@@ -6925,6 +6925,23 @@ static int si_dpm_enable(struct amdgpu_device *adev)
 	return 0;
 }
 
+static int si_set_temperature_range(struct amdgpu_device *adev)
+{
+	int ret;
+
+	ret = si_thermal_enable_alert(adev, false);
+	if (ret)
+		return ret;
+	ret = si_thermal_set_temperature_range(adev, R600_TEMP_RANGE_MIN, R600_TEMP_RANGE_MAX);
+	if (ret)
+		return ret;
+	ret = si_thermal_enable_alert(adev, true);
+	if (ret)
+		return ret;
+
+	return ret;
+}
+
 static void si_dpm_disable(struct amdgpu_device *adev)
 {
 	struct rv7xx_power_info *pi = rv770_get_pi(adev);
@@ -7608,6 +7625,18 @@ static int si_dpm_process_interrupt(struct amdgpu_device *adev,
 
 static int si_dpm_late_init(void *handle)
 {
+	int ret;
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	if (!adev->pm.dpm_enabled)
+		return 0;
+
+	ret = si_set_temperature_range(adev);
+	if (ret)
+		return ret;
+#if 0 //TODO ?
+	si_dpm_powergate_uvd(adev, true);
+#endif
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
index 4cd43bbec910..1d96eb274d72 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
@@ -1303,13 +1303,12 @@ static int arcturus_get_power_limit(struct smu_context *smu,
 	if (default_power_limit)
 		*default_power_limit = power_limit;
 
-	if (smu->od_enabled) {
+	if (smu->od_enabled)
 		od_percent_upper = le32_to_cpu(powerplay_table->overdrive_table.max[SMU_11_0_ODSETTING_POWERPERCENTAGE]);
-		od_percent_lower = le32_to_cpu(powerplay_table->overdrive_table.min[SMU_11_0_ODSETTING_POWERPERCENTAGE]);
-	} else {
+	else
 		od_percent_upper = 0;
-		od_percent_lower = 100;
-	}
+
+	od_percent_lower = le32_to_cpu(powerplay_table->overdrive_table.min[SMU_11_0_ODSETTING_POWERPERCENTAGE]);
 
 	dev_dbg(smu->adev->dev, "od percent upper:%d, od percent lower:%d (default power: %d)\n",
 							od_percent_upper, od_percent_lower, power_limit);
@@ -2273,8 +2272,8 @@ static uint16_t arcturus_get_current_pcie_link_speed(struct smu_context *smu)
 
 	/* TODO: confirm this on real target */
 	esm_ctrl = RREG32_PCIE(smnPCIE_ESM_CTRL);
-	if ((esm_ctrl >> 15) & 0x1FFFF)
-		return (uint16_t)(((esm_ctrl >> 8) & 0x3F) + 128);
+	if ((esm_ctrl >> 15) & 0x1)
+		return (uint16_t)(((esm_ctrl >> 8) & 0x7F) + 128);
 
 	return smu_v11_0_get_current_pcie_link_speed(smu);
 }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c
index 8d1d29ffb0f1..ed189a3878eb 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c
@@ -2357,13 +2357,12 @@ static int navi10_get_power_limit(struct smu_context *smu,
 		*default_power_limit = power_limit;
 
 	if (smu->od_enabled &&
-		    navi10_od_feature_is_supported(od_settings, SMU_11_0_ODCAP_POWER_LIMIT)) {
+		    navi10_od_feature_is_supported(od_settings, SMU_11_0_ODCAP_POWER_LIMIT))
 		od_percent_upper = le32_to_cpu(powerplay_table->overdrive_table.max[SMU_11_0_ODSETTING_POWERPERCENTAGE]);
-		od_percent_lower = le32_to_cpu(powerplay_table->overdrive_table.min[SMU_11_0_ODSETTING_POWERPERCENTAGE]);
-	} else {
+	else
 		od_percent_upper = 0;
-		od_percent_lower = 100;
-	}
+
+	od_percent_lower = le32_to_cpu(powerplay_table->overdrive_table.min[SMU_11_0_ODSETTING_POWERPERCENTAGE]);
 
 	dev_dbg(smu->adev->dev, "od percent upper:%d, od percent lower:%d (default power: %d)\n",
 					od_percent_upper, od_percent_lower, power_limit);
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
index 21fc033528fa..e2ad2b972ab0 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
@@ -640,13 +640,12 @@ static int sienna_cichlid_get_power_limit(struct smu_context *smu,
 	if (default_power_limit)
 		*default_power_limit = power_limit;
 
-	if (smu->od_enabled) {
+	if (smu->od_enabled)
 		od_percent_upper = le32_to_cpu(powerplay_table->overdrive_table.max[SMU_11_0_7_ODSETTING_POWERPERCENTAGE]);
-		od_percent_lower = le32_to_cpu(powerplay_table->overdrive_table.min[SMU_11_0_7_ODSETTING_POWERPERCENTAGE]);
-	} else {
+	else
 		od_percent_upper = 0;
-		od_percent_lower = 100;
-	}
+
+	od_percent_lower = le32_to_cpu(powerplay_table->overdrive_table.min[SMU_11_0_7_ODSETTING_POWERPERCENTAGE]);
 
 	dev_dbg(smu->adev->dev, "od percent upper:%d, od percent lower:%d (default power: %d)\n",
 					od_percent_upper, od_percent_lower, power_limit);
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
index dd9bcbd630a1..f41ac6465f2a 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
@@ -1682,8 +1682,8 @@ static int aldebaran_get_current_pcie_link_speed(struct smu_context *smu)
 
 	/* TODO: confirm this on real target */
 	esm_ctrl = RREG32_PCIE(smnPCIE_ESM_CTRL);
-	if ((esm_ctrl >> 15) & 0x1FFFF)
-		return (((esm_ctrl >> 8) & 0x3F) + 128);
+	if ((esm_ctrl >> 15) & 0x1)
+		return (((esm_ctrl >> 8) & 0x7F) + 128);
 
 	return smu_v13_0_get_current_pcie_link_speed(smu);
 }
@@ -1746,10 +1746,12 @@ static ssize_t aldebaran_get_gpu_metrics(struct smu_context *smu,
 
 	gpu_metrics->current_fan_speed = 0;
 
-	gpu_metrics->pcie_link_width =
-		smu_v13_0_get_current_pcie_link_width(smu);
-	gpu_metrics->pcie_link_speed =
-		aldebaran_get_current_pcie_link_speed(smu);
+	if (!amdgpu_sriov_vf(smu->adev)) {
+		gpu_metrics->pcie_link_width =
+			smu_v13_0_get_current_pcie_link_width(smu);
+		gpu_metrics->pcie_link_speed =
+			aldebaran_get_current_pcie_link_speed(smu);
+	}
 
 	gpu_metrics->system_clock_counter = ktime_get_boottime_ns();
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
index a9954ffc02c5..9b80f18ea6c3 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
@@ -2369,13 +2369,12 @@ static int smu_v13_0_0_get_power_limit(struct smu_context *smu,
 	if (default_power_limit)
 		*default_power_limit = power_limit;
 
-	if (smu->od_enabled) {
+	if (smu->od_enabled)
 		od_percent_upper = le32_to_cpu(powerplay_table->overdrive_table.max[SMU_13_0_0_ODSETTING_POWERPERCENTAGE]);
-		od_percent_lower = le32_to_cpu(powerplay_table->overdrive_table.min[SMU_13_0_0_ODSETTING_POWERPERCENTAGE]);
-	} else {
+	else
 		od_percent_upper = 0;
-		od_percent_lower = 100;
-	}
+
+	od_percent_lower = le32_to_cpu(powerplay_table->overdrive_table.min[SMU_13_0_0_ODSETTING_POWERPERCENTAGE]);
 
 	dev_dbg(smu->adev->dev, "od percent upper:%d, od percent lower:%d (default power: %d)\n",
 					od_percent_upper, od_percent_lower, power_limit);
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 45a84fd5dc04..3957af057d54 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -2147,8 +2147,8 @@ static int smu_v13_0_6_get_current_pcie_link_speed(struct smu_context *smu)
 
 	/* TODO: confirm this on real target */
 	esm_ctrl = RREG32_PCIE(smnPCIE_ESM_CTRL);
-	if ((esm_ctrl >> 15) & 0x1FFFF)
-		return (((esm_ctrl >> 8) & 0x3F) + 128);
+	if ((esm_ctrl >> 15) & 0x1)
+		return (((esm_ctrl >> 8) & 0x7F) + 128);
 
 	speed_level = (RREG32_PCIE(smnPCIE_LC_SPEED_CNTL) &
 		PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK)
@@ -2228,14 +2228,16 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table
 	gpu_metrics->gfxclk_lock_status = GET_METRIC_FIELD(GfxLockXCDMak) >> GET_INST(GC, 0);
 
 	if (!(adev->flags & AMD_IS_APU)) {
-		link_width_level = smu_v13_0_6_get_current_pcie_link_width_level(smu);
-		if (link_width_level > MAX_LINK_WIDTH)
-			link_width_level = 0;
-
-		gpu_metrics->pcie_link_width =
-			DECODE_LANE_WIDTH(link_width_level);
-		gpu_metrics->pcie_link_speed =
-			smu_v13_0_6_get_current_pcie_link_speed(smu);
+		if (!amdgpu_sriov_vf(adev)) {
+			link_width_level = smu_v13_0_6_get_current_pcie_link_width_level(smu);
+			if (link_width_level > MAX_LINK_WIDTH)
+				link_width_level = 0;
+
+			gpu_metrics->pcie_link_width =
+				DECODE_LANE_WIDTH(link_width_level);
+			gpu_metrics->pcie_link_speed =
+				smu_v13_0_6_get_current_pcie_link_speed(smu);
+		}
 		gpu_metrics->pcie_bandwidth_acc =
 				SMUQ10_ROUND(metrics_x->PcieBandwidthAcc[0]);
 		gpu_metrics->pcie_bandwidth_inst =
@@ -2306,8 +2308,8 @@ static int smu_v13_0_6_mode2_reset(struct smu_context *smu)
 	ret = smu_cmn_send_msg_without_waiting(smu, (uint16_t)index,
 					       SMU_RESET_MODE_2);
 
-	/* This is similar to FLR, wait till max FLR timeout */
-	msleep(100);
+	/* Reset takes a bit longer, wait for 200ms. */
+	msleep(200);
 
 	dev_dbg(smu->adev->dev, "restore config space...\n");
 	/* Restore the config space saved during init */
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
index 0ffdb58af74e..3dc7b60cb075 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c
@@ -2333,13 +2333,12 @@ static int smu_v13_0_7_get_power_limit(struct smu_context *smu,
 	if (default_power_limit)
 		*default_power_limit = power_limit;
 
-	if (smu->od_enabled) {
+	if (smu->od_enabled)
 		od_percent_upper = le32_to_cpu(powerplay_table->overdrive_table.max[SMU_13_0_7_ODSETTING_POWERPERCENTAGE]);
-		od_percent_lower = le32_to_cpu(powerplay_table->overdrive_table.min[SMU_13_0_7_ODSETTING_POWERPERCENTAGE]);
-	} else {
+	else
 		od_percent_upper = 0;
-		od_percent_lower = 100;
-	}
+
+	od_percent_lower = le32_to_cpu(powerplay_table->overdrive_table.min[SMU_13_0_7_ODSETTING_POWERPERCENTAGE]);
 
 	dev_dbg(smu->adev->dev, "od percent upper:%d, od percent lower:%d (default power: %d)\n",
 					od_percent_upper, od_percent_lower, power_limit);
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c
index 2aa7e9945a0b..7ac9bc0df8fd 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c
@@ -229,8 +229,6 @@ int smu_v14_0_check_fw_version(struct smu_context *smu)
 		smu->smc_driver_if_version = SMU14_DRIVER_IF_VERSION_SMU_V14_0_2;
 		break;
 	case IP_VERSION(14, 0, 0):
-		if ((smu->smc_fw_version < 0x5d3a00))
-			dev_warn(smu->adev->dev, "The PMFW version(%x) is behind in this BIOS!\n", smu->smc_fw_version);
 		smu->smc_driver_if_version = SMU14_DRIVER_IF_VERSION_SMU_V14_0_0;
 		break;
 	default:
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c
index 47fdbae4adfc..9310c4758e38 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c
@@ -261,7 +261,10 @@ static int smu_v14_0_0_get_smu_metrics_data(struct smu_context *smu,
 		*value = metrics->MpipuclkFrequency;
 		break;
 	case METRICS_AVERAGE_GFXACTIVITY:
-		*value = metrics->GfxActivity / 100;
+		if ((smu->smc_fw_version > 0x5d4600))
+			*value = metrics->GfxActivity;
+		else
+			*value = metrics->GfxActivity / 100;
 		break;
 	case METRICS_AVERAGE_VCNACTIVITY:
 		*value = metrics->VcnActivity / 100;